CLASSIC

Consistent Longitudinal Alignment and Segmentation for Serial Image Computing

Zhong Xue, Dinggang Shen, Christos Davatzikos

Research output: Contribution to journalArticle

Abstract

This paper proposes a temporally consistent and spatially adaptive longitudinal MR brain image segmentation algorithm, referred to as CLASSIC, which aims at obtaining accurate measurements of rates of change of regional and global brain volumes from serial MR images. The algorithm incorporates image-adaptive clustering, spatiotemporal smoothness constraints, and image warping to jointly segment a series of 3-D MR brain images of the same subject that might be undergoing changes due to development, aging, or disease. Morphological changes, such as growth or atrophy, are also estimated as part of the algorithm. Experimental results on simulated and real longitudinal MR brain images show both segmentation accuracy and longitudinal consistency.

Original languageEnglish (US)
Pages (from-to)388-399
Number of pages12
JournalNeuroImage
Volume30
Issue number2
DOIs
StatePublished - Apr 1 2006
Externally publishedYes

Fingerprint

Brain
Atrophy
Cluster Analysis
Growth

Keywords

  • Brain atrophy
  • Brain growth
  • Fuzzy clustering
  • Image segmentation
  • Longitudinal brain image analysis
  • Serial scans
  • Volumetry

ASJC Scopus subject areas

  • Cognitive Neuroscience
  • Neurology

Cite this

CLASSIC : Consistent Longitudinal Alignment and Segmentation for Serial Image Computing. / Xue, Zhong; Shen, Dinggang; Davatzikos, Christos.

In: NeuroImage, Vol. 30, No. 2, 01.04.2006, p. 388-399.

Research output: Contribution to journalArticle

Xue, Zhong ; Shen, Dinggang ; Davatzikos, Christos. / CLASSIC : Consistent Longitudinal Alignment and Segmentation for Serial Image Computing. In: NeuroImage. 2006 ; Vol. 30, No. 2. pp. 388-399.
@article{65fe35c8b3d745fd9b01b4b43bc9e7f9,
title = "CLASSIC: Consistent Longitudinal Alignment and Segmentation for Serial Image Computing",
abstract = "This paper proposes a temporally consistent and spatially adaptive longitudinal MR brain image segmentation algorithm, referred to as CLASSIC, which aims at obtaining accurate measurements of rates of change of regional and global brain volumes from serial MR images. The algorithm incorporates image-adaptive clustering, spatiotemporal smoothness constraints, and image warping to jointly segment a series of 3-D MR brain images of the same subject that might be undergoing changes due to development, aging, or disease. Morphological changes, such as growth or atrophy, are also estimated as part of the algorithm. Experimental results on simulated and real longitudinal MR brain images show both segmentation accuracy and longitudinal consistency.",
keywords = "Brain atrophy, Brain growth, Fuzzy clustering, Image segmentation, Longitudinal brain image analysis, Serial scans, Volumetry",
author = "Zhong Xue and Dinggang Shen and Christos Davatzikos",
year = "2006",
month = "4",
day = "1",
doi = "10.1016/j.neuroimage.2005.09.054",
language = "English (US)",
volume = "30",
pages = "388--399",
journal = "NeuroImage",
issn = "1053-8119",
publisher = "Academic Press Inc.",
number = "2",

}

TY - JOUR

T1 - CLASSIC

T2 - Consistent Longitudinal Alignment and Segmentation for Serial Image Computing

AU - Xue, Zhong

AU - Shen, Dinggang

AU - Davatzikos, Christos

PY - 2006/4/1

Y1 - 2006/4/1

N2 - This paper proposes a temporally consistent and spatially adaptive longitudinal MR brain image segmentation algorithm, referred to as CLASSIC, which aims at obtaining accurate measurements of rates of change of regional and global brain volumes from serial MR images. The algorithm incorporates image-adaptive clustering, spatiotemporal smoothness constraints, and image warping to jointly segment a series of 3-D MR brain images of the same subject that might be undergoing changes due to development, aging, or disease. Morphological changes, such as growth or atrophy, are also estimated as part of the algorithm. Experimental results on simulated and real longitudinal MR brain images show both segmentation accuracy and longitudinal consistency.

AB - This paper proposes a temporally consistent and spatially adaptive longitudinal MR brain image segmentation algorithm, referred to as CLASSIC, which aims at obtaining accurate measurements of rates of change of regional and global brain volumes from serial MR images. The algorithm incorporates image-adaptive clustering, spatiotemporal smoothness constraints, and image warping to jointly segment a series of 3-D MR brain images of the same subject that might be undergoing changes due to development, aging, or disease. Morphological changes, such as growth or atrophy, are also estimated as part of the algorithm. Experimental results on simulated and real longitudinal MR brain images show both segmentation accuracy and longitudinal consistency.

KW - Brain atrophy

KW - Brain growth

KW - Fuzzy clustering

KW - Image segmentation

KW - Longitudinal brain image analysis

KW - Serial scans

KW - Volumetry

UR - http://www.scopus.com/inward/record.url?scp=33645115816&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=33645115816&partnerID=8YFLogxK

U2 - 10.1016/j.neuroimage.2005.09.054

DO - 10.1016/j.neuroimage.2005.09.054

M3 - Article

VL - 30

SP - 388

EP - 399

JO - NeuroImage

JF - NeuroImage

SN - 1053-8119

IS - 2

ER -